[FFmpeg-devel] [PATCH v6] libswscale/ppc: VSX-optimize 9-16 bit yuv2planeX

Lauri Kasanen cand at gmx.com
Sun Jan 27 10:31:42 EET 2019


On Mon, 14 Jan 2019 16:13:52 +0100
Michael Niedermayer <michael at niedermayer.cc> wrote:

> On Sun, Jan 13, 2019 at 10:26:20AM +0200, Lauri Kasanen wrote:
> > ./ffmpeg_g -f rawvideo -pix_fmt rgb24 -s hd1080 -i /dev/zero -pix_fmt yuv420p16be \
> > -s 1920x1728 -f null -vframes 100 -v error -nostats -
> > 
> > 9-14 bit funcs get about 6x speedup, 16-bit gets about 15x.
> > Fate passes, each format tested with an image to video conversion.
> > 
> > Only POWER8 includes 32-bit vector multiplies, so POWER7 is locked out
> > of the 16-bit function. This includes the vec_mulo/mule functions too,
> > not just vmuluwm.
...
> > v6: No patch changes, updated bench numbers without skips.
> 
> fate does not get worse from this patch on qemu ppc32be and ppc64le 

Ping

- Lauri


More information about the ffmpeg-devel mailing list